The prediction of road users’ future motion is a critical task in supporting advanced driver-assistance systems (ADAS). It plays an even more crucial role for autonomous driving (AD) in enabling the planning and execution of safe driving maneuvers. Based on our previous work, Context-Aware Scene Prediction Network (CASPNet), an improved system, CASPNet++, is proposed. In this work, we focus on further enhancing the interaction modeling and scene understanding to support the joint prediction of all road users in a scene using spatiotemporal grids to model future occupancy. Moreover, an instance-based output head is introduced to provide multimodal trajectories for agents of interest. In extensive quantitative and qualitative analysis, we demonstrate the scalability of CASPNet++ in utilizing and fusing diverse environmental input sources such as HD maps, Radar detection, and Lidar segmentation. Tested on the urban-focused prediction dataset nuScenes, CASPNet++ reaches state-of-the-art performance. The model has been deployed in a testing vehicle, running in real-time at 20 Hz with moderate computational resources alongside a machine learning-based perception system.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    CASPNet++: Joint Multi-Agent Motion Prediction


    Beteiligte:
    Schafer, Maximilian (Autor:in) / Zhao, Kun (Autor:in) / Kummert, Anton (Autor:in)


    Erscheinungsdatum :

    02.06.2024


    Format / Umfang :

    2540357 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    MAP-Former: Multi-Agent-Pair Gaussian Joint Prediction

    Steiner, Marlon / Klemp, Marvin / Stiller, Christoph | IEEE | 2024


    Goal-Oriented Multi-Modal Motion Prediction for Multi-Agent Systems

    James, Anegi / Bakolas, Efstathios | AIAA | 2025



    Multi-agent trajectory prediction

    NARAYANAN SRIRAM NOCHUR / LIU BUYU / MOSLEMI RAMIN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    MULTI-AGENT TRAJECTORY PREDICTION

    NARAYANAN SRIRAM NOCHUR / LIU BUYU / MOSLEMI RAMIN et al. | Europäisches Patentamt | 2021

    Freier Zugriff